HILDA-household-stats-for-hhs-retiree-super-exceeds-1.6-million

HILDA-household-stats-for-hhs-retiree-super-exceeds-1,600,000.R

library(testthat)
library(dplyr)
library(dtplyr)
library(data.table)
library(hildaData)
library(hildaExtra)
library(sih.hilda.1314)
library(taxstats)
library(magrittr)

write_clipboard <- function(tbl){
  write.table(tbl, "clipboard", sep = "\t", row.names = FALSE)
}

wave14 <- as.data.table(Combined_n140c)

enumerate_weights_by_hhid <-
  wave14 %>%
  select(xwaveid, 
         id_hh = nhhrhid, 
         wt_enumerated = nhhwte) %>%
  .[, .(n_persons = sum(wt_enumerated)), keyby = id_hh]

enumerate_weight_by_xwaveid <-
  wave14 %>%
  select(xwaveid,
         wt_enumerated = nhhwte)

retirees_super_balances <-
  wave14 %>%
  select(xwaveid,
         id_hh = nhhrhid,
         Age = nhgage,
         Super_balance_retirees = npwsupri,
         Super_balance_workers = npwsupwi) %>%
  mutate(Super_balance = Super_balance_retirees + Super_balance_workers) %>%
  select(-Super_balance_retirees, -Super_balance_workers) %>%
  .[(Age >= 65)] %>%
  # To see households together
  .[order(id_hh)]

retirees_with_balances_over <- function(balance) {
  retirees_super_balances[Super_balance > balance] %>%
    enumerate_weight_by_xwaveid[., on = "xwaveid"]
}

net_wealth_by_idhh <-
  svy_id_wealthPartition %>%
  .[survey == "hilda"] %>%
  .[, .(net_wealth = sum(asset_value)), keyby = id_hh]

gross_home_value_by_idhh <-
  wave14 %>%
  select(id_hh = nhhrhid,
         home_value = nhsvalui) %>%
  # refer only to person
  setkey(id_hh) %>%
  unique(by = key(.))

net_home_value_by_idhh <-
  svy_id_wealthPartition %>%
  .[survey == "hilda"] %>%
  .[asset_type == "Owner-occupied housing (net of liabilities)"] %>%
  select(id_hh, net_home_value = asset_value) %>%
  setkey(id_hh)

# Check of number of persons
test_that("Number of persons in HILDA within benchmark (15%) of 2013-14 sample file", {
  n_HILDA <-
    retirees_with_balances_over(1.0e6) %>%
    filter(Age >= 65) %$%
    sum(wt_enumerated)
  
  n_ATO <-
    sample_file_1314 %>%
    copy %>%
    merge(age_range_decoder, by = "age_range") %>%
    filter(age_range_description >= "65 to 69") %$%
    sum(MCS_Ttl_Acnt_Bal > 1.0e6) * 50
  
  expect_lte(abs(1 - n_HILDA / n_ATO), 0.15)
})


retirees_with_balances_over(1.6e6) %>%
  setkey(id_hh) %>%
  net_wealth_by_idhh[.] %>%
  net_home_value_by_idhh[.] %>%
  gross_home_value_by_idhh[.] %>%
  .[, .(
    avg_home_value = weighted.mean(home_value, wt_enumerated),
    n_persons = sum(wt_enumerated),
    avg_net_home_value = weighted.mean(net_home_value, wt_enumerated),
    avg_hh_net_wealth = weighted.mean(net_wealth, wt_enumerated)
  )] %>%
  write_clipboard

retirees_with_balances_over(1.3e6) %>%
  setkey(id_hh) %>%
  net_wealth_by_idhh[.] %>%
  net_home_value_by_idhh[.] %>%
  gross_home_value_by_idhh[.] %>%
  .[, .(
    avg_home_value = weighted.mean(home_value, wt_enumerated),
    n_persons = sum(wt_enumerated),
    avg_net_home_value = weighted.mean(net_home_value, wt_enumerated),
    avg_hh_net_wealth = weighted.mean(net_wealth, wt_enumerated)
  )] %>%
  write_clipboard

retirees_with_balances_over(-Inf) %>%
  setkey(id_hh) %>%
  net_wealth_by_idhh[.] %>%
  net_home_value_by_idhh[.] %>%
  gross_home_value_by_idhh[.] %>%
  .[, .(
    avg_home_value = weighted.mean(home_value, wt_enumerated),
    n_persons = sum(wt_enumerated),
    avg_net_home_value = weighted.mean(net_home_value, wt_enumerated),
    avg_hh_net_wealth = weighted.mean(net_wealth, wt_enumerated)
  )] %>%
  write_clipboard